Livestock Research for Rural Development 32 (4) 2020 LRRD Search LRRD Misssion Guide for preparation of papers LRRD Newsletter

Citation of this paper

Advancing climate smart agriculture: adoption potential of multiple on-farm dairy production strategies among farmers in Murang’a County, Kenya

N C Maindi, I M Osuga1 and M G Gicheha1

Department of Agricultural and Resource Economics, Jomo Kenyatta University of Agriculture and Technology, P O Box 62000-00200, Nairobi, Kenya
1 Department of Animal Sciences, Jomo Kenyatta University of Agriculture and Technology, P O Box 62000-00200, Nairobi, Kenya
isaac.osuga@jkuat.ac.ke

Abstract

Improving uptake and intensification of farm-level utilisation of climate smart agriculture (CSA) strategies among farmers is essential to develop resilient livestock production systems for sustainable livelihoods and food security while addressing climate change adaptation and mitigation. However, empirical basis for sector-specific understanding of the adoption behaviour of the farmers to climate change is merely established, prompting the current study on adoption of CSA strategies among resource-constrained dairy farmers particularly in Sub Saharan Africa (SSA). The case study employed a joint analysis framework of both multivariate probit (MVP) and ordered probit models to analyse farmers’ joint adoption decisions for four on-farm strategies namely: improved fodder, feed conservation, artificial insemination and manure management.

The case study, involving 132 dairy farmers from a representative Murang’a County in Kenya revealed that adoption of CSA practices among farmers is widespread in the study area, with majority of the farmers (87%) adopting at least two of the four considered strategies. However, the specific adoption rates were relatively low, ranging from 27% for manure management to 66% for improved fodder and thus the need to target the less adopted strategies and incentivise the farmers to intensify their implementation. The findings show interdependence of the strategies with complementarity and substitutionality relationships among the practices. The interdependence can facilitate the tailoring of suitable packages of strategies which are interrelated to optimise their synergies. Capital, gender, water availability, market access and infrastructure and social networks were found to be the most important determinants of adoption decision as well as the intensity of adoption. These findings from dairy sector-specific data in Kenya are significant to provide tailored and targeted policies in scaling up adoption and intensification of CSA strategies to advance climate smart dairy production systems in SSA.

Keywords: climate change, mitigation, multivariate probit model, ordered probit model, sustainable dairying


Introduction

Globally, human population is expected to increase from the current 7.7 to 9.7 billion by the year 2050 with double increases occurring in low income countries particularly in SSA (UN 2019). This population growth coupled with urbanization, increase in incomes and shifts in dietary patterns is expected to double the demand for animal protein by 60% from the current production of 8.5 billion tonnes/year (FAO 2018). Already, SSA constitutes a quarter of the world’s 821 million food insecure people (FAO 2018) and 78% of the world’s extremely poor population (AfDB 2019). This region has employed production intensification interventions to enhance market oriented livestock production systems among the smallholder farmers to meet the growing demand and concurrently improve livelihoods and food security (Bosire et al 2019). However, due to the resource intensive nature of the livestock products, the quest to improve productivity through intensification may put enormous pressure on the already depleted resources particularly land and water (Thornton and Herrero 2014). More so, intensification efforts often encounter the negative impacts of climate change (Herrero et al 2016). Equally, the livestock sector accounts for 14.5% of the total global greenhouse gas (GHG) emissions, contributing further to climate change (Gerber et al 2013). Thus, the SSA economies face a constant trade-off between agricultural development and reducing agricultural activities to mitigate GHG (IPCC 2019). Thus, the main concern is how to improve livestock productivity and take advantage of the opportunities of rising demand for livestock products while safeguarding the livestock and environmental footprint by mitigating climate change.

Livestock is a source of food and nutritional security to an estimated 820 million food insecure people providing 13% of the total food calories and 33% of proteins worldwide (Gerber et al 2013). In SSA smallholder mixed crop-livestock systems, livestock employs about 1.3 million people (Herrero et al 2016), provides manure for soil fertilization to about 60% of the world’s cropping area (Notenbaert et al 2017) and insurance in the event of crop failure (Enahoro et al 2019; Shinbrot et al 2019). In this region, livestock contributes to national economic growth. For instance, the dairy industry in Kenya, which is the largest in SSA, contributes 8% of Gross Domestic Product (GDP) (Erickssen and Crane 2018). Nevertheless, the mixed dairy systems contribute the highest GHG emissions in the SSA and at the same time the sector is highly vulnerable to drought and climate change (Gerber et al 2013). The sector, like the rest of livestock production systems in SSA, is characterised by low productivity, which is attributed to warmer climate, poor animal genetics and health, feed shortages, resource degradation, poor infrastructure, institutions and markets and most importantly high vulnerability to risks and impacts of climate change (Erickssen and Crane 2018; Teklewold et al 2017; Thornton and Herrero 2014).

The historical and anticipated long term changes in rainfall intensity and variability, increased frequency and severity of floods, droughts and extreme temperatures are expected to significantly increase livestock productivity risks and impacts (Ericksen and Crane 2018; Escarcha et al 2018). Climate change affects feed intake and conversion efficiency, rates of weight gain, reproductive and lactation performance, diseases and pest infestations, heat stress, carrying capacities, quantity and quality of feeds and water as well as morbidity and mortality rates (Seo 2015; Nardone et al 2010; Michalk et al 2018; Thornton and Cramer 2012). The effects results in pronounced economic losses, market instability and developmental setbacks in safeguarding food and nutritional security across the world (Michalk et al 2018; IPCC 2019). The climatic effects on livestock varies by region, production system and animal breed (Notenbaert et al 2017). For instance, Henry et al (2018) reported lower milk production of 40% to 60% in Holstein-Jersey dairy breed in tropical conditions. The impacts are projected to disproportionately impact SSA countries because their economies depend on agriculture. For instance, in Kenya, the United States Agency for International Development (USAID) report projects that the climate variability and extremes will cost the country a GDP loss equivalent to 2.6 per annum by 2030 (USAID 2018). Furthermore, due to widespread poverty and low investment incentives in this region, the smallholders have limited capacities to adapt, build resilience, cope and recover from the impacts of climate shocks (Teklewold et al 2017; Thornton and Herrero 2014), thus exacerbating poverty, food security and economic decline (Kabubo-Mariara and Mulwa 2019; Shinbrot et al 2019).

The production response in the face of progressive climate change underpins climate smart agriculture (CSA) to simultaneously mitigate the intensities of GHG, improve climate resilience and on-farm productivity to address poverty and food insecurity (FAO 2010). In Kenya, the National Climate Change and Response Strategy of 2010 recognise and promote adaptation and mitigation measures to transform agriculture to a low carbon and climate resilient sector (GOK 2017). The on-farm adaptation and mitigation strategies among the dairy farmers include improving animal genetics, feed conservation, management of grasslands, production of improved pastures and fodders, herd size reduction and manure management (Erickssen and Crane 2018; Khatri-Chhetri et al 2017; Thornton and Cramer 2012; Zhang et al 2017). The CSA practices among resource poor famers may reduce GHG emissions (Hristov et al 2013), enhance climate change resilience (Gerber et al 2013), agricultural productivity (Notenbaert et al 2017), farm income (Teklewold et al 2017) and food security (Wekesa et al 2018). Despite the benefits and continued support from national and transnational institutions, the uptake of the CSA practices among the farmers is weak and uneven (Kpadonou et al 2016; Wekesa et al 2018). Thus, to enhance adoption of CSA, it calls for a sector-specific understanding on the determinants of farmer’s adoption decisions and choices of practices. The sector-specific data would provide effective and informed policy and practice and target investments to scale up adoption.

Farmers adopt a portfolio of CSA strategies simultaneously to deal with the multiple and overlapping agricultural risks and constraints and exploit their co-benefits as supplements, substitutes and complements (Aryal et al 2018; Hailemariam et al 2019). Given choices and trade-offs when making adoption decisions, farmers take into account the synergies among the practices (Adego et al 2019; Kpadonou et al 2017). Thus ignoring such interdependence may fail to account for the trade-offs and synergies among the practices and this may yield to biased and inconsistent estimates. Furthermore, failure to account for the CSA paradigm of triple-win objectives of adaptation, mitigation and food security may limit the scope of policy implications (Aryal et al 2018; Hailemariam et al 2019; Kpadonou et al 2017). Empirical studies on the adoption of CSA employing joint analytical framework focuses mainly on soil and water conservation practices (Aryal et al 2018; Hailemariam et al 2019; Ouédraogo et al 2019; Shinbrot et al 2019; Teklewold et al 2017; Wekesa et al 2018) and mixed crop and livestock practices (Descheemaeker et al 2016; Kpadonou et al 2017; Khatri-Chhetri et al 2017). The studies focussing on CSA adoption specifically in livestock systems employ, descriptive, qualitative and inferential static tools (Erickssen and Crane 2018; Escarcha et al 2018; Notenbaert et al 2017; Rojas-Downing et al 2017; Zhang et al 2019). Moreover, literature barely reveals studies on dairy sub sector.

In line with this, the recent reviews (Henry et al 218) and studies (Erickssen and Crane 2018; Feleke et al 2016) recommend for a deeper understanding of the incentives and barriers of CSA adoption as part of expansive and varied contexts and sectors so as to inform on transformation and reorientation of SSA agricultural systems to climate change. This study, therefore sought to jointly analyse the drivers and incentives that accelerate or constrain decisions to adopt multiple CSA practices and assess the intensity of adoption in Kenya.


Methodology

Study Area and Data

The study was conducted in Murang’a County, in Central region of Kenya. Geographically, the county is located between the latitude of 0 o 34′ and 10o 7′S and longitude of 36o 37o 27′ East, the county occupies an area of 2,5558 Km 2. Murang’a County lies between 914 meters above sea level (ASL) in the East and 3,353m ASL to the West along the slopes of Aberdare Mountains. The county is characterised by three climatic regions; the western region is wet and humid owing to its proximity to Mt. Kenya and Aberdare Ranges while the central region and western regions have subtropical and semi-arid conditions respectively. The county receives bimodal rainfall where the long rains fall from March to May and the short rains in the months of October and November.

Like the rest of the world, impacts of climate change in the Murang’a are evident as indicated by rainfall variability, high temperatures and extreme occurrences of prolonged drought and floods resulting to landslides, soil erosion and infrastructural damages, drying of rivers, and declining agricultural productivity. The county anchors its economy in agriculture. An estimated 80% of the residents practice cash cropping (coffee and tea, avocadoes), food crop farming (bananas, potatoes and other horticultural crops) and livestock farming particularly dairy farming which is most prominent in the county. The area is relevant for the study because of its prominence in dairy farming, the local farmer’s long experience with CSA practices, the commonality of the observed practices and socioeconomic characteristics in most arid and semi-arid areas of Africa.

The cross-sectional data used for this study was derived from farm household survey in the year 2015/2016. The sample for the study was purposefully drawn from Kandara sub-County based on its high potential for dairy production in the entire Murang’a County. A multistage sampling technique was employed in the first stage to select two villages per ward from the six wards (Ng'araria, Ithiru, Muruka, Kagundu-Ini, Ruchu and Gaichanjiru) in the study area. The villages were selected to ensure that the sampling frame is a representative of the local typology of agro-ecological zones and dairy farming systems in the study area. In the second stage, 11 dairy farming households were randomly selected from each village totalling to 132 farm households. With the aid of a well-structured and pre-tested questionnaire, on-farm face to face interviews were administered by well-trained enumerators. After the inconsistent responses and missing data were discarded, 112 observations were used for analysis.

Econometric model specifications
Multivariate Probit

The choice of the empirical specification for the study is grounded on the climate adaptation literature (Adego et al 2019; Aryal et al 2018; Feleke et al 2016; Hailemariam et al 2019; Kpadonou et al 2017; Teklewold et al 2017). In this study, the farmers’ decisions to select a combination of strategies to address the multiple challenges they face (such as soil degradation, drought, poor productivity, limited resources, feed shortages and nutrient deficiency) and exploit their triple win co-benefits. Whereas a farmer may adopt a combination of CSA, depending on the constraints and benefits of the strategies, the decision to adopt a strategy may be conditioned by the choice of other strategies due to their interdependence either as complements (positive correlation) or as substitutes (negative correlation). Employing univariate probit or logit to estimate the adoption decisions of the farmers may yield biased conclusions because they fail to capture the potential correlations (interdependence of the practices) of unobserved disturbances between different adoption equations (Greene 2003). The suitability of multinomial probit is limited due to endogeneity problem such that it could lead to difficulties in making interpretations for the simultaneous effect of explanatory variables on specific outcome variables. In this context, the MVP is suitable because it jointly allows for modelling of multiple CSA adoption decisions while allowing for unobservable disturbances to be correlated freely.

A farmer is more likely to adopt a specific CSA strategy if the expected utility of its adoption is higher than non-adoption. For instance the ith farm household (i = 1,..,…N) is facing a decision on whether to adopt the jth practice or not (Where jth denotes the choice of improved fodder (IF) (napier grass and/or desmodium), artificial insemination (AI), feed conservation (FC) (silage) and manure management (MM)). Let represent the benefits associated with the adoption of jth practice and otherwise. The farmer will choose to adopt the jth practice on the farm if Y*ipj U*j - Uo > 0. The net utility from adoption of jth practice is a latent variable determined by observed household and farm factors (X´ip) and the normally distributed error term (ip).

Applying the indicator function, the unobserved binary choices in equation (1) for each of the CSA practices can be given as follows:

Where if Yi j is a dichotomous observable variable indicating the adoption decision of the ith farmer as regards to the jth practice. In the MVP, where adoption of multiple practices is possible, the error terms simultaneously follow a multivariate normal distribution (MVN) with a mean of zero and variance normalized to unity, i.e. (UIF, UAI, UFC, UMM ) ~ MVN (0, Ω) and symmetric covariance matrix given as follows:

Where p(rho) denotes the pair-wise correlation coefficient of the error terms of the adoption equations to be analysed in the model. In the model, if the value of the off-diagonal elements in the covariance matrix is equal to zero, then it justifies the application of MVP in the place of univariate probit or logit models. The sign and significance of the correlation coefficients (p)signifies the nature of the relationship between adoption equations (Kpadonou et al 2017). A positive correlation is interpreted to mean complementarity and a negative correlation signifies subsitutionality.

Ordered probit model

The MVP considers the propensity of adoption of a particular SCA practice without a distinction on the number of practices adopted by the farmer, i.e. those who adopt a single practice and those who adopt multiple practices. In this context, the explanatory variables that influence the propensity of adoption may differ as regards to the number of practices adopted. Understanding of factors that affect the intensity of CSA adoption is relevant to promote multiple adoptions of technologies to jointly achieve multiple objectives. Following Kpadonou al (2017), Hailemariam et al (2019) and Aryal et al (2018), the study applied ordered probit approach to model the determinants of intensity of adoption of CSA practices. In this case, intensity of adoption is a count variable measured by the number of practices adopted in an individual farm as a dependent count variable; which takes a value ranging from 0 to 4. Count data is usually analysed employing Poisson regression model with the underlying assumption that all events have the same probability of occurrence. However, in this study, the likelihood of adopting a practice could differ from the propensity of adopting the second one, third or fourth one; and with the adoption of the latter practices, the farmer already has had exposure to information, gained experience and benefits about the practice. Furthermore, of importance is that, the constraints associated with the CSA practices and the determinants of their adoption may vary from one practice to another, thus differentiated probabilities of adoption of each practice. Inherent to this, the study considered the number of CSA practices adopted as an ordinal variable, hence the application of ordered probit model, which is expressed as follows:

Where Zip ϵ {,…..n}, is the categorical outcome variable denoting the number of CSA practices adopted by the farmer on the farm, . Xip is a vector of exogenous household and farm factors and β is a vector of parameters to be estimated and ip random disturbances which follows a normal distribution with a mean of zero and unit variance. The values of Zip can be observed and hence the probabilities of each outcome can be expressed as:

Following the recommendations of Greene (2003), the marginal effects of each outcome have been calculated for easy interpretation.

Dependent variables

This study considered four best-bet CSA strategies which were selected based on the data availability, commonality of drivers of their adoption and their potential to deliver triple win objectives that seeks to jointly improve climate resilience and productivity for income and food security while minimising contribution of GHG from the dairy sector (Erickssen and Crane 2018; Feleke et al 2016; Notenbaert et al 2017). The strategies include: improved fodder, artificial insemination, feed conservation and manure management. The strategies complement each other, for instance manure management provides soil nutrients leading to increased productivity of fodder crops; manure also improves soil moisture content which increases the resilience of the fodder crops to drought. The strategies also provide synergies to deliver co-benefits of the triple win objectives as shown in Table 1.

Table 1. Selection of CSA strategies and their contribution to triple win objectives

CSA strategy

Definition and justification for inclusion

Improved
fodder

Definition

Cultivation of high yielding, high protein and digestible genetically improved perennial pastures, fodders and legume varieties. Helps address seasonality of feeds and deficiencies in quantity and quality of basal diets (Zhang et al 2017).

Adaptation

Drought tolerance; increased resilience through deep rooted that can absorb water and nutrients from deep soil layers (Thornton and Herrero 2014)

Mitigation

Nitrogen fixing of legumes reduce emissions from fertilizer use in soils; reduction of methane emissions from enteric fermentation of cattle through improved digestion (Notenbaert et al 2017).

Productivity

Improved digestibility results to increased dry matter intake (DMI), high energy levels and thus improved productivity (dairy weight gain, reproductive and lactation performance) (Notenbaert et al 2017; Erickssen and Crane 2018).

Household welfare

Enhanced incomes from improved productivity, better market price and better production efficiency (Thornton and Herrero 2014).

Artificial
Insemination
(AI)

Definition

Improvement of animal genetics for targeted traits to improve productive, reproductive traits (Thornton and Herrero 2014)

Adaptation

Increased resilience to heat stress tolerance and pest and disease resistance (Henry et al 2018).

Mitigation

Reduced GHG emissions per unit of product through improved herd health, reproductive performance, herd reduction through improvement of few quality animal stocks (Gerber et al 2013).

Productivity

Improved herd health, growth rates, fertility rates, age at first calving and reduced mortality rates (Rojas-Downing et al 2017; Notenbaert et al 2017).

Household welfare

Improved resource use efficiency and productive efficiency and thus enhanced incomes and food security (Erickssen and Crane 2018).

Feed
conservation

Definition

Preservation and storage of forages in form of hay or silage for utilisation during periods of shortages (Thornton and Herrero 2014).

Adaptation

Availability of feed during shortages.

Mitigation

Improved quality of low basal diets reduces the GHG through enteric fermentation ( Khatri-Chhetri et al 2017).

Productivity

Improved DMI and digestibility and resource use efficiency thus enhanced productivity for income and food security (Notenbaert et al 2017; Zhang et al 2017)

Manure
management

Definition

Collection, accumulation, storage processing (composting) and applying to the soil/ direct application to pastures.

Adaptation

Improvement of degraded soils

Mitigation

Reduction of methane and nitrous oxide emissions from manure; reduction of nitrogen loss in soils and fossil fuels (replaced by biogas) (Hristov et al 2013).

Productivity

Improvements in farm productivity of pasture, feed and food crops (Notenbaert et al 2017).

Household welfare

Enhanced income and food security due to enhanced productivity; Production of biogas; substitute for fossil fuels

Source : Author

Independent variables and Hypotheses

The choice of explanatory variables included in the study is based on theoretical frameworks and past climate adaptation literature (Adego et al 2019; Erickssen and Crane 2018; Garcia et al 2016; Hailemariam et al 2019; Issahaku and Abdulai 2019; Wekesa et al 2018). Specifically, the models analysed household demographics as proxies for human capital. These include age, education and gender of the household head and household size. In the study, it is hypothesised that household heads with formal education are more likely to adopt CSA strategies due to their enhanced capacity to receive, process and understand relevant information necessary to make adoption decisions (Gbetibou et al 2010). As regards the age of the household head, these studies take indeterminate direction because older individuals, on one hand, may have accumulated physical and social capita as well as have more experience in farming and climate change. On the other hand, older people are associated with limited horizons in planning, physical deterioration and perceived as more risk averse (Aryal et al 2018). As regards to farm factors, given the fixed information and transaction costs associated with adoption of CSA strategies, the small sized farms may be disadvantaged and thus not likely to adopt (Gbetibouo, et al 2010; Kpadonou et al 2017). The farm characteristics included in this study are farm size, number of dairy cattle the farmer owns, the value of the dairy cattle quantified in Kenya shillings (KES), the value of the tank the farm owns, well, availability of piped water and irrigation equipment in the farm.

Most of the CSA strategies demand a lot of inputs and thus the farmers’ endowments in wealth and assets may stimulate CSA uptake. The proxies used in the study for the farmer’s wealth and assets include income from milk, aggregate farm income, income from off farm engagement, value of total assets the farmer owns and access to credit. However, off farm engagement may limit household’s total time spent in farming and farm labour requirements (Feleke et al 2016). Social capital facilitates information exchange and asset accumulation which my drive adoption of CSA practices (Hailemariam et al 2019). The proxies included for the variable are household membership in agriculture related group, trust in the extension services and community kinsmen. In the face of asymmetric information and high transaction costs, the study also hypothesised that access to information and other institutions such as input and output markets, extension services increased the likelihood of CSA adoption. The dummy ward variable was also included to control for the fixed sub county effects to reflect any specific institutional arrangements, which may affect the capacity of the farmers to adopt CSA in a particular ward. The detailed description of the variables used in the study is presented in Table 2.

Table 2. Description and descriptive statistics of the variables used in the study

Variables

Variable Description

Mean

SD

Expected Sign

Dependent variables

Improved fodder

1if improved fodder is produced;

0.66

0.48

Artificial Insemination

1 if AI is practised; 0 otherwise

0.44

0.50

Feed conservation

1 if fodder conservation (hay, silage)

0.29

0.46

Manure management

1 if manure management is practiced

0.27

0.45

Independent variables

Household demographics

Age

Age of the household head

64.54

10.02

±

Gender

1If the household head is male

0.79

0.41

+

Educ

Number of years of formal education

9.01

3.71

+

Hhsize

Number of household members > 18 years

2.38

0.91

±

Farm factors

Farmsize

Farm size (acres)

2.02

1.51

+

Dairycattle

Number of dairy cattle

3.08

1.75

+

Dairycattlevalue

Log value of dairy cattle (KES)

4.91

0.31

+

Tankvalue

Log value of tank in (KES)

3.29

2.00

+

Well

1 if farm has well; 0, otherwise

0.30

0.46

+

Pipewater

1 if farm has piped water; 0, otherwise

0.68

0.47

+

Irrigneqip

1 if farmer owns irrigation equipment

0.22

0.42

+

Wealth and assets

Milkincome

Log milk income in (KES)

4.92

0.45

+

Farmincome

Log total farm income in (KES)

5.63

0.39

+

Offfarminc

1 if farmer has off farm income

0.77

0.42

-

Assets

Log value of total assets (KES)

5.76

0.43

+

Credit

1 if farmer has access to credit;

0.21

0.41

+

Infrastructure, extension and market access

Inputmktdst

Distance from farm to input markets (km)

3.09

2.22

-

Outputmktdst

Distance from farm to output markets (km)

2.44

1.40

-

Distmotor

Distance from farm to motor able road (km)

0.44

0.45

-

Disttarmac

Distance from farm to tarmac road (km)

1.47

1.21

-

Distextn

Distance from farm to extension office ((km)

2.60

1.41

-

Distai

Distance to artificial insemination centre (km)

2.22

1.32

-

Infoclimate

Number of Information sources on climate

4.29

2.64

+

Social capital networks and social conformity

Membagric

1 if a membership in agricultural related group

0.78

0.42

+

Trustextn

Trust index of the extension service providers

1.91

0.15

+

Trustcommty

Index of trust to the community

1.80

1.30

+

Note: SD = standard deviation; KES=Kenyan shillings; km= kilometres


Results and discussion

Adoption rates and intensity of adoption

Adoption rates were higher in improved fodder (66%) closely followed by AI (44%) whereas the rates were relatively low for feed conservation and manure management, which were estimated at 29% and 27% respectively (Table 2). The number of dairy CSA strategies adopted across the households is presented in Table 3. As shown, the intensity of adoption ranged between 0 to 4 practices. It can be observed that most of the farmers (47.3%) used at least one practice. The results further reveal that an estimated 38% of the farmers used two practices, 9% used three practices and 5% used four practices. A finding that is consistent with Kpadonou et al (2017) that adoption of CSA practices vary across socio-economic settings of the households and the types of practices.

Table 3. Intensity of adoption of CSA practices

Number of
practices

Percentage of
the adopters

Commutated
percentages

0

0.9

2.7

1

47.3

48.2

2

38.4

86.6

3

8.9

95.5

4

4.5

100

Total

100

100

The findings reveal that there is still necessity to upscale efforts to increase uptake of CSA practices in an effort to enhance farmer’s incomes and food security and resilience to climate change while mitigating GHG emissions and degradation of resources in Kenya. As such, understanding the major drivers and constraints to adoption and intensity of adoption of CSA practices is crucial to provide evidence-based policy making for agricultural development in SSA.

Interdependence of the practices

The simultaneous adoption of a number of CSA practices in the study area indicates a possibility of correlation (interdependence) between the CSA choices. In the study, employing pair-wise correlation across the residuals of the MVP, the estimates are presented in Table 4. As indicated, the likelihood ratio test (Chi2 (6) = 11. 1221; Prob > chi2 = 0.000) rejects the null hypothesis of zero covariance of the error terms across the equations. The finding thus justifies the application of MVP approach as explained in the section 2.2.1 of empirical model specification. Of the 6 pairs of SCA practices, 4 pairwise correlations between the error terms were statistically significant. The findings confirm that the adoption decisions of CSA practices are interdependent due to not only substitutability or complementarity in dairy farming, but also potentially due to omitted factors influencing all possible adoption decisions. This result to farmers’ making decisions not to adopt a single practice because ideally but the probability of adopting a practice is dependent of whether other CSA practices have already been adopted.

Table 4. Correlation matrix of the error terms

Rho1

Rho2

Rho3

Rho4

Rho1

1

-0.054

0.187**

0.260*

Rho2

1

-0.135

0.265*

Rho3

1

0.219*

Rho4

1

Likelihood ratio test of: rho =21 =rho 31= rho41= rho42=rho43=0. Chi2 (6) = 11.2994;
Prob > X 2 =000. *p < 0.1; **p < 0.05; *** p <0.01
1 = feed conservation; 2= manure management; 3 = Artificial insemination; 4 = improved fodder; negative sign = substitutability relationship; Positive sign = complementarity relationship

Feed conservation and AI measures were positively interdependent, implying that farmers often combine the two practices to complement each other. The finding is consistent with Erickssen and Crane (2018) that in the face of dwindling resources, AI results to reduction of animal stocking rates through selection of few high-performance animals thus milk production gains can be herd constant with the fewer animals. Besides, the finding underscores the importance of feed conservation in complementing AI in an effort to improve feed resource efficiency and production efficiency. Further the positive correlation between feed conservation and AI implies that farmers complemented improved fodder pastures with AI to address the effects of droughts and accrue synergies of drought tolerance and resilience of both livestock and livestock feeds as well as optimisation of milk productivity (Zhang et al 2017).

It can be observed that feed conservation complemented improved fodder practice, suggesting that farmers adopted the two practices simultaneously to help address seasonality and deficiencies in quantity and quality of basal diets as well as exploit the co-benefits of improved DMI and digestibility (Notenbaert et al 2017). The use improved pastures such as napier grass and legumes usually leads to reduction of methane production in the rumen, which makes the practice climate smart especially under tropical conditions (Maselema and Chigwa 2017). Although there was no significant correlation found between manure management and AI, the two practices substitute each other to improve production efficiency as well as minimise GHG emissions (Gerber et al 2013). The results further indicate a positive association between manure management and improved fodder, suggesting complementarity among the two practices that could result to co-benefits. The finding corroborate with on-farm observations and Kpadonou et al (2017) that compost or direct application of manure to the fodder or pasture crops improved productivity by effecting soil fertilization and water retention capacity.

Determinants of adoption of CSA strategies

The parameter estimates of the MVP regression are presented in Table 5. The MVP fits the data well as indicated by the Wald test, X 2(104) = 171.06; Prob > X2 = 0.000, thus rejecting the null hypothesis that all regression coefficients in the model equations are jointly equal to zero. This also indicates the relevance of MVP to account for the unobserved (interdependence) correlations across CSA multiple adoption decisions. In regard to household characteristics and in consistent with Mulwa et al (2017), gender of the household head significantly and positively affects adoption of CSA particularly improved fodder and feed conservation practices. This may be attributed to access to productive resources such as land, labour and assets unlike the female headed households. Consistent with other findings (Feleke et al 2017; Gbetobou et al 2010), and diverging from the study of Hailemariam et al (2019), age and education of the household head were insignificant in all the four practices. As indicated, the negative coefficient signs imply a decreasing likelihood of adopting the practices. Although educated people are presumed to be well informed and risk averse, in the study area, most of the educated individuals are engaged in off farm activities. On the other hand, older people are associated with limited horizons in planning, physical deterioration and perceived as more risk averse thus are less likely to adapt CSA practices (Aryal et al 2018).

In corroboration with Aryal et al (2018) and Feleke et al (2017) on farm characteristics, the results show that increase in the number of livestock units leads to an increase in the likelihood of adoption of improved fodder and feed conservation practices. The results further indicate that households with better breeds of cattle (quantified by the value of the cattle in KES), are more likely to adapt AI in addition to improved fodder and feed conservation practices. The stronger effect of dairy cattle and dairy cattle value on feed conservation is consistent with the expectations, reflecting and underscoring the importance of the practice in addressing the shortages of feeds during periods of drought. Moreover, the combination of the three practices may result to co-benefits in an effort to cope and offset the negative effects of climate change on both livestock and feed resources. Ownership of a Well had a positive and significant impact on adoption of improved fodder. This finding is consistent with Aryal et al (2018) that underscore the importance of availability and accessibility to water resources in uptake of adoption and adjustments in agricultural systems on and beyond the farm level. In this finding, thus, availability and accessibility of water is a necessary condition for adopting improved fodder due to intense frequency and quantity of water required for their growth and development.

Table 5. Parameter estimates of adoption of dairy climate smart practices: multivariate probit model

Independent Variables

Dependent variables

Improved fodder

Artificial insemination

Feed conservation

Manure processing

Coeff.

SE

Coeff.

SE

Coeff.

SE

Coeff.

SE

Household factors

Age

-0.326

0.008

-0.012

0.014

-0.016

0.014

-0.019

0 .016

Educ

-0.667

0.034

-0.009

0.050

0.086

0.052

0.021

0.058

Hhsize

-0.300

0.025

0.166

0.181

0.147

0.225

0.149

0.164

Gender

0.799

0.033**

0.693

0.377

0.694

0.355*

0.268

0.421

Farm factors

Farmsize

0.129

0.118

0.170

0.116

0.131

0.116

-0.138

0.161

Dairycattle

0.577

0.232**

0.070

0.102

0.507

0.130***

0.001

0.118

Dairycattleval

0.907

0.041**

1.100

0.536**

2.685

0.848***

0.245

0.755

Tankvalue

0.248

0.161

-0.086

0.097

-0.031

0.109

0.119

0.099

Well

0.617

0.285**

0.137

0.313

0.251

0.299

0.128

0 .343

Pipewater

0.563

0.043

-0.159

0.378

0.235

0.450

0.310

0.404

Irrigneqip

0.664

0.624

-0.302

0.402

0.443

0.509

-0.669

0 .561

Wealth and assets

Milkincome

1.612

0.724**

0.089

0.367

1.011

0.364***

-0.357

0.437

Farmincome

-1.497

0.714**

0.952

0.434**

0.270

0.458

-0.486

0.513

Offfarminc

0.152

0.732

-0.683

0.335**

0.203

0.466

-0.835

0.400**

Assets

0.053

0.925

1.310

0.510**

0.367

0.540

0.431

0.518

Credit

0.300

0.437

0.398

0.362

0.326

0.359

0 .958

0.420**

Infrastructure, extension and market access

Inputmktdst

-0.065

0.144

-0.393

0.127***

-0.298

0.098***

0.039

0.086

Outputmktdst

0.378

0.208

0.434

0.141***

-0.098

0.134

0.382

0.129***

Disttarmac

0.524

0.215**

-0.143

0.153***

0.205

0.149

-0.507

0.225**

Distmotor

1.463

0.536***

1.005

0.364

-0.421

0.381

0.571

0.463

Distextn

0.146

0.147

0.088

0.119

0.130

0.122

-0.566

0.543

Distai

0.283

0.144**

0.107

0.121

-0.303

0.128**

0.290

0.152

Infoclimate

0.040

0.072

0.018

0.056

0.105

0.054*

0.040

0.072

Social capital networks & social conformity

Agricmemb

0.567

0.310

0.509

0.216**

0.344

0.239

0.064

0.218

Trustextn

0.555

1.412

0.672

0.126

0.509

0.162

0.136

0.407

Trustcommty

0.033

0.016**

0.019

0.009**

0.008

0.011

0.010

0.011

Warddummy

0.124

0.661

0.348

0.160**

0.022

0.885

0.285

0.070

Constant

-1.498

0.250*

-6.112

4.271

-2.727

0.658***

-1.088

0.855

Likelihood ratio test of: rho =21 =rho 31= rho41= rho42=rho43=0. Chi2 (6) = 11.2994; Prob > X 2 =000 *p < 0.1; **p < 0.05; *** p <0.01

Farm income had a positive and significant effect on improved fodder and improved breeding practices but not on feed conservation and manure management practices. This may be attributed to the economic implications involved in obtaining the improved fodder seedlings and AI services. An increase in assets is associated with adoption of improved breeding. Asset endowed and higher-income farmers tend to be wealthier and risk averse, have more access to information, have a longer-term horizon in their planning and lower discount rates compared to less-income farmers (Gbetibouo, et al 2010). In this finding the choice of improved fodder and AI are perceived as long-term horizon strategies adopted in mitigating, adopting and enhancing tolerance of both livestock and feeds to the persistent detrimental effects of climate change. The two practices provide synergies which can lead to maximising of milk productivity (Descheemaeker et al 2016; Rojas-Downing et al 2017). The off-farm income decreased the likelihood of adoption of improved breeding and manure processing. The finding upholds the studies of Feleke et al (2017) and Issahaku and Abdulai (2019) that alludes that off-farm engagement and adoption of CSA practices may compete for time and household labour thus leading to loss of labour. Farmers with access to credit are more likely to adopt manure processing but have a low likelihood of adopting the other three CSA practices. Adoption of long term technologies and innovations requires initial investment in form of either own or borrowed capital to purchase inputs and labour requirements, thus access to credit and/or borrowing capacity removes cash constraints thus greatly enhancing the uptake of CSA (Mulwa et al 2017).

Distance to input and output market significantly increased the likelihood of adopting CSA strategies. Access to markets determines the availability of inputs required for climate change adoption, a means of exchanging and sharing information and experiences on CSA among farmers. However, the accessibility of the markets depends on the condition of infrastructure. The findings show that distance to the tarmacked road was significant for improved fodder, animal breeding and manure processing. Comparably, distance to motorable roads increases the likelihood of adoption of improved fodder. Distance to AI service providers had a positive effect on adoption of improved fodder and interestingly it was not significant for adoption of animal breeding underscoring the importance of improving livestock genetics in the study area. Similar to findings of Hailemariam et al (2019) and Aryal et al (2018), infrastructure reduces transaction costs, facilitates transport of farm inputs and outputs, opportunity costs of time and acquiring timely production, market and climate information thus enhancing the capacity of adoption of CSA practices. Access to information was significant at 10% significance level for feed conservation. Although not significant for other three practices, the finding upholds the results of Mulwa et al (2017) and Issahaku and Abdulai (2019) that exposure of farmers to climate information increases the knowledge and awareness of climate change and CSA skills and practices and hence the likelihood of the farmers to take measures to cope and mitigate the risks associated with climate change. On the social capital networks and social conformity, membership in at least one agricultural group and trust in fellow community members had a positive effect on the choice of CSA practices. Such social networks and social conformity facilitates information flows about markets, opportunities, potential risks and shocks of climate change and CSA as well as better access to markets and financial services (Hailemariam et al 2019). Social capital networks also provide informal insurance systems in times of crisis as well as reciprocity incentives which provides incentives to the farmers to adopt CSA strategies they particularly perceive to possess positive productivity benefits.

Determinants of intensity of CSA adoption

In the preceding section, the study analysed factors influencing the farmer’s decision to adopt a specific CSA practice interdependent of one or more of other practices. But this does not permit for an understanding of the determinants of farmers’ choices on adoption of multiple practices. Evidence shows that joint adoption of multiple practices generates synergies and numerous benefits in optimising of stable and improved productivity, enhanced livelihoods and ecosystem gains (Kabubo-Mariara and Mulwa 2019; Wekesa et al 2018). In this section, the study analysed the factors influencing intensity of adoption of CSA practices, with a proxy of the number of practices adopted. Table 6 presents the estimated coefficients and marginal effects of the explanatory variables on each dependant variable, in this case the intensity of adoption. The effects of independent variables are highly varied over different levels of CSA adoption intensity. The coefficient for age was significantly negative, suggesting that older farmers are less likely to intensify the use of CSA practices. The gender of the household head was significant and positive, implying that female-headed households have a high likelihood of intensifying the implementation of CSA, a finding that is in line with Kpadonou et al (2017) with the assumption that women have limited control of resources such as land, labour and financial services as well as time availability.

Holding all the factors constant, an increase in the size of the landholding increases the propensity of adopting one, two, three or four practices by 7.7%, 28.5%, 36.3% and 23.6% respectively.

Table 6. Estimates of the ordered probit model and marginal effects of independent variables

Variables

Marginal effects

Coefficients

Pr(Y=1|X)

Pr(Y=2|X)

Pr(Y=3|X)

Pr(Y=4|X)

Age

-0.022

0.002

0.037

0.171

0.436

(0.042)**

(0.662)

(0.506)

(0.236)

(0.000)***

Gender

0.160

0.074

0.278

0.365

0.244

(0.571)

(0.007)***

(0.000)***

(0.000)***

(0.000)***

Farmsize

-0.004

0.078

0.285

0.364

0.236

(0.970)

(0.010)***

(0.000)***

(0.000)***

(0.000)***

Dairycattle

0.238

0.178

0.380

0.306

0.124

(0.000)***

(0.001)***

(0.000)***

( 0.000)***

(0.004)

Dairycattleval

1.104

0.163

0.371

0.316

0.136

(0.006)***

(0.004)***

(0.000)***

(0.000)***

(0.003)***

Well

0.684

0.029

0.176

0.347

0.353

(0.005)***

(0.094)

(0.001)***

(0.000)***

(0.000)***

Farmincome

-1.057

0.134

0.766

0.868

0.001

(0.000 )***

(0.911)

(0.896)

(0.872)

(0.838)

Offfarminc

0.053

0.077

0.283

0.364

0.238

(0.841)

(0.006)***

(0.000)***

(0.000)***

(0.000)***

Assets

1.081

0.064

0.000

0.801

0.441

(0.000)***

(0.731)

(0.859)

(0.889)

(0.907)

Inputmktdst

-0.146

0.043

0.215

0.361

0.062

( 0.006)***

(0.038)**

(0.000)***

(0.000)***

(0.000)***

Outputmktdst

0.321

0.169

0.375

0.312

0.130

(0.001)***

(0.001)***

(0.000)***

(0.000)***

(0.003)***

Disttarmac

-0.034

0.076

0.282

0.364

0.239

(0.729)

(0.006)***

(0.000)***

(0.000)***

(0.000)***

Distmotor

1.069

0.022

0.150

0.331

0.379

(0.000)***

(0.099)

(0.001)***

(0.000)***

(0.000)***

Distai

0.206

0.048

0.226

0.364

0.299

( 0.046)**

(0.037)**

(0.000)

(0.000)

(0.000)

Agricmemb

0.382

0.118

0.335

0.345

0.180

( 0.004)***

(0.008)***

(0.000)***

(0.000)***

(0.001)***

Trustextn

0.008

0.227

0.399

0.272

0.093

(0.248)

(0.210)

(0.000)***

(0.034)***

(0.312)

Trustcommty

0.003

0.110

0.328

0.350

0.188

(0.630)

(0.179)

(0.001)***

(0.000)***

(0.074)

Observations

112

Wald X2(25)

104.07

Prob X2

0.000***

Log likelihood

-147.476

Note: Other non-significant variables include Educ, household size, Tankvalue, Pipewater, irrignequip, milkincome, Credit, Distextn, Infoclimate Figures in parenthesis are robust standard errors. *p < 0.1; **p < 0.05; *** p < 0.01

The numbers of dairy cattle and dairy cattle value have positive association with the intensity of adoption of CSA practices. The findings further show that ownership of a Well had a positive effect on the number of CSA practices adopted, implying that availability and accessibility of water can be a constraint to intensifying coping and mitigating responses to climate change in dairy farming. In regards to off farm income, the findings indicate that it has positive and significant effect in adopting multiple practices but with varying degrees of marginal probabilities. In line with Aryal et al (2018), distance to the markets, tarmac and motorable roads in addition to AI services had a positive impact on the number of practices. Social capital in respect to membership in agricultural collectives, trust in community kinsmen increases the marginal probability of adopting more than two practices by 39.8% and 32.8% respectively. Social capital facilitates to prepare farmers for climate risks and vulnerabilities, providing timely measures to cope, mitigate, adopt as well as alleviate the consequences of climate variability (Hailemariam et al 2019). Finally, trust in extension service providers helps to raise awareness, disseminate information, knowledge, skills and benefits of CSA practices, thus the variable has a significant positive effect on intensity of adoption.


Conclusion and policy implications

The study found that adoption of CSA practices among farmers is widespread in the study area; however, the intensity of adoption of the specific practices is very low. About 5% use all the four practices measured, implying a huge potential to improve and upscale adoption rates of the specific practices. There is need for policy makers to target practices that are less adopted and incentivise the farmers in order to intensify their implementation. The findings also reveal complementarity and subsitutionality among the practices, implying that policy change affecting one practice may have spill over effects on the adoption of other interrelated practices. Thus, this interdependence can facilitate to tailor suitable packages of strategies interrelated to optimise their synergies for more climate smart dairy production systems.

The findings underscore the capital intensive nature of CSA strategies as well as access to water source. There is need to incentivise and take effective measures to improve access to credit to significantly promote the farmer’s abilities and capacities to adopt and intensify the use of CSA. The CSA practices being long-term investments, the policy makers need to target the female headed households who exhibited low incentives in uptake and intensification in the utilisation of CSA, probably because of their limited access and control of land, labour and financial resources. There is need for policy makers to focus and consider water accessibility and availability to heighten uptake of CSA in order to enhance the resilience of the farmers and help to counteract the adverse effects of climate change and variability. Given the significance of information on climate in uptake of CSA, it is important to create and disseminate more awareness, knowledge and information on climate change and its corresponding adoption mechanisms to increase uptake and intensification of utilisation of CSA. Finally, the significant role played by social capital networks has policy implications particularly on not only establishing but strengthening the local institutional arrangements to accelerate uptake of CSA.


Acknowledgement

The authors acknowledge the role of Ministry of Agriculture, Livestock, Fisheries and Irrigation in Murang’a County, enumerators and farmers in Murang’a County, Kenya.


References

Adego T, Simane B and Woldie G A 2019 The impact of adaptation practices on crop productivity in northwest Ethiopia: an endogenous switching estimation. Development Studies Research, 6(1), 129-141. doi:10.1080/21665095.2019.1678186

African Development Bank (AfDB) 2019 African Economic Outlook. URL: https://www.afdb.org/en/knowledge/publications/african-economic-outlook

Aryal J P, Rahut D B, Maharjan S and Erenstein O 2018 Factors affecting the adoption of multiple climate-smart agricultural practices in the Indo-Gangetic Plains of India. Natural Resources Forum, 42(3), 141-158. doi:10.1111/1477-8947.12152

Bosire C K, Rao E J O, Muchenje V, Van Wijk M, Ogutu J O, Mekonnen M M, Auma J O, Lukuyu B and Hammond J 2019 Adaptation opportunities for smallholder dairy farmers facing resource scarcity: Integrated livestock, water and land management. Agriculture, Ecosystems & Environment, 284, 106592. doi:10.1016/j.agee.2019.106592

Descheemaeker K, Oosting S J, Homann-Kee T S, Masikati P, Falconnier G N and Giller K E 2016 Climate change adaptation and mitigation in smallholder crop–livestock systems in sub-Saharan Africa: a call for integrated impact assessments. Regional Environmental Change, 16(8), 2331-2343. doi:10.1007/s10113-016-0957-8

Enahoro D, Njuri N, Thornton P and Staal S J 2019 A review of projections of demand and supply of livestock-derived foods and the implications for livestock sector management in LSIL focus countries. Mid-Project Research Report of the Feed the Future Innovation Lab for Livestock Systems (LSIL) Futures Foresight Component, Module I (Quantitative Scenario Modelling). CCAFS Working Paper no. 262 Wageningen, the Netherlands: CGIAR Research Program on Climate Change, Agriculture and Food Security (CCAFS). Available online at: www.ccafs.cgiar.org

Ericksen P and Crane T 2018 The feasibility of low emissions development interventions for the East African livestock sector: Lessons from Kenya and Ethiopia. ILRI Research Report 46. Nairobi, Kenya: International Livestock Research Institute (ILRI). Available from: https://ccafs.cgiar.org/publications/feasibility-low-emissions-development-interventions-east-african-livestock-sector#.XmObZ0pRU2w

Escarcha J, Lassa J and Zander K 2018 Livestock Under Climate Change: A Systematic Review of Impacts and Adaptation. Climate, 6(3), 54. doi:10.3390/cli6030054

FAO 2010 Climate Smart Agriculture: Policies, Practices and Financing For Food Security, Adaptation and Mitigation. Food and Agriculture Organization of the United Nations (FAO), Rome, Italy. Available from: http://www.fao.org/3/i1881e/i1881e00.htm

FAO 2018 Agricultural Outlook 2018-2027, Dairy and Dairy Products. 163-174. doi:10.1787/agr_outlook-2018-10-en

Feleke F B, Berhe M, Gebru G and Hoag D 2016 Determinants of adaptation choices to climate change by sheep and goat farmers in Northern Ethiopia: the case of Southern and Central Tigray, Ethiopia. SpringerPlus, 5(1). doi:10.1186/s40064-016-3042-3

Garcia J S, Silvestri S and Barnes A P 2016 The potential for adoption of climate smart agricultural practices in Sub-Saharan livestock systems. Regional Environmental Change. ISSN 1436-3798.

Gbetibouo G A, Hassan R M and Ringler C 2010 Modelling farmers' adaptation strategies for climate change and variability: The case of the Limpopo Basin, South Africa. Agrekon, 49(2), 217-234. doi:10.1080/03031853.2010.491294

Gerber P J, Steinfeld H, Henderson B, Mottet A, Opio C, Dijkman J, Falcucci A and Tempio G 2013 Tackling climate change through livestock – A global assessment of emissions and mitigation opportunities. Food and Agriculture Organization of the United Nations (FAO), Rome Available from: http://www.fao.org/3/i3437e/i3437e.pdf

Government of Kenya (GOK) 2017 Kenya Climate Smart Agriculture Strategy-2017-2026. Retrieved from: https://www.adaptation-undp.org/resources/plans-and-policies-relevance-naps-least-developed-countries-ldcs/kenya-climate-smart

Hailemariam T, Alemu M and Gunnar K 2019 Climate change adaptation: a study of multiple climate-smart practices in the Nile Basin of Ethiopia, Climate and Development, 11:2, 180-192, DOI: 10.1080/17565529.2018.1442801

Henry B K, Eckard R J and Beauchemin K A 2018 Review: Adaptation of ruminant livestock production systems to climate changes. animal, 12(2), 445-456. doi:10.1017/s1751731118001301

Herrero M, Henderson B, Havlík P, Thornton P K, Conant R T, Smith P, Wirsenius S, Hristov A N, Gerber P, Gill M, Butterbach-Bahl K, Valin H, Garnett T, Stehfest E 2016 Greenhouse gas mitigation potentials in the livestock sector. Nature Climate Change 6, 452–461 http://dx.doi.org/10.1038/NCLIMATE2925

Hristov A N, Oh J, Lee C, Meinen R, Montes F, Ott T, Firkins J, Rotz A, Dell C, Adesogan A, Yang W, Tricarico J, Kebreab E, Waghorn G, Dijkstra J and Oosting S 2013 Mitigation of greenhouse gas emissions in livestock production – A review of technical options for non-CO2 emissions. Edited by Pierre J. Gerber, Benjamin Henderson and Harinder P.S. Makkar. FAO Animal Production and Health Paper No. 177. FAO, Rome, Italy.

Intergovernmental Panel on Climate Change (IPCC) 2019 AR6 Climate Change 2021: Impacts, Adaptation and Vulnerability. Retrieved from: https://www.ipcc.ch/report/sixth-assessment-report-working-group-ii/

Issahaku G and Abdulai A 2019 Adoption of climate-smart practices and its impact on farm performance and risk exposure among smallholder farmers in Ghana. Australian Journal of Agricultural and Resource Economics. doi:10.1111/1467-8489.12357

Kabubo-Mariara J and Mulwa R 2019 Adaptation to climate change and climate variability and its implications for household food security in Kenya. Food Security. doi:10.1007/s12571-019-00965-4

Khatri-Chhetri A, Aggarwal P, Joshi P and Vyas S 2017 Farmers' prioritization of climate-smart agriculture (CSA) technologies. Agricultural Systems, 151, 184-191. doi:10.1016/j.agsy.2016.10.005

Kpadonou R A, Owiyo T, Barbier B, Denton F, Rutabingwa F and Kiema A 2017 Advancing climate-smart-agriculture in developing drylands: Joint analysis of the adoption of multiple on-farm soil and water conservation technologies in West African Sahel. Land Use Policy, 61, 196-207. doi:10.1016/j.landusepol.2016.10.050

Maselema D and Chigwa F C 2017 The potential of Richardia scabra and fodder tree leaf meals in reducing enteric methane from dairy cows during dry season. Livestock Research for Rural Development. Volume 29, Article #51. Retrieved March 7, 2020, from http://www.lrrd.org/lrrd29/3/mase29051.html

Michalk D L, Kemp D R, Badgery W B, Wu J, Zhang Y and Thomassin P J 2018 Sustainability and future food security-A global perspective for livestock production. Land Degradation & Development, 30(5), 561-573. doi:10.1002/ldr.3217

Mulwa C, Marenya P, Rahut D B and Kassie M 2017 Response to climate risks among smallholder farmers in Malawi: A multivariate probit assessment of the role of information, household demographics, and farm characteristics. Climate Risk Management, 16, 208-221. doi:10.1016/j.crm.2017.01.002

Nardone A, Ronchi B, Lacetera N, Ranieri M S and Bernabucci U 2010 Effects of climate change on animal production and sustainability of livestock systems. Livestock Science 130, 57–69. doi: https://doi.org/10.1016/j.livsci.2010.02.011

Notenbaert A, Pfeifer C, Silvestri S and Herrero M 2017 Targeting, out-scaling and prioritising climate-smart interventions in agricultural systems: Lessons from applying a generic framework to the livestock sector in sub-Saharan Africa. Agricultural Systems, 151, 153-162. doi:10.1016/j.agsy.2016.05.017

Ouédraogo M, Houessionon P, Zougmoré R B and Partey S T 2019 Uptake of Climate-Smart Agricultural Technologies and Practices: Actual and Potential Adoption Rates in the Climate-Smart Village Site of Mali. Sustainability, 11(17), 4710. doi:10.3390/su11174710

Rojas-Downing M M, Nejadhashemi A P, Harrigan T and Woznicki S A 2017 Climate change and livestock: Impacts, adaptation, and mitigation. Climate Risk Management, 16, 145-163. doi:10.1016/j.crm.2017.02.001

Seo S N 2015 Adapting to extreme climates: Raising animals in hot and arid ecosystems in Australia. International Journal of Biometeorology 59, 541–550 https://doi.org/10.1007/s00484-014-0867-8

Shinbrot X A, Jones K W, Rivera-Castañeda A, López-Báez W and Ojima D S 2019 Smallholder Farmer Adoption of Climate-Related Adaptation Strategies: The Importance of Vulnerability Context, Livelihood Assets, and Climate Perceptions. Environmental Management, 63(5), 583-595. doi:10.1007/s00267-019-01152-z

Teklewold H, Mekonnen A, Kohlin G and Di Falco S 2017 Does Adoption of Multiple Climate-Smart Practices Improve Farmers’ Climate Resilience? Empirical Evidence from the Nile Basin of Ethiopia. Climate Change Economics, 08(01), 1750001. doi:10.1142/s2010007817500014

Thornton P and Cramer L (eds) 2012 Impacts of climate change on the agricultural and aquatic systems and natural resources within the CGIAR’s mandate. CCAFS Working Paper 23. CGIAR Research Program on Climate Change, Agriculture and Food Security (CCAFS). Copenhagen, Denmark. Available online at: www.ccafs.cgiar.org

Thornton P K and Herrero M 2014 Climate change adaptation in mixed crop–livestock systems in developing countries. Global Food Security 3, 99–107. doi:10.1016/j.gfs.2014.02.002

United Nations, Department of Economic and Social Affairs, Population Division 2019 World Population Prospects 2019: Highlights (ST/ESA/SER.A/423)

USAID 2018 Climate Risk Profile Kenya. https://www.climatelinks.org/sites/default/files/asset/document/2018_USAID-ATLAS-Project_Climate-Risk-Profile-Kenya.pdf

Wekesa B M, Ayuya O I and Lagat J K 2018 Effect of climate-smart agricultural practices on household food security in smallholder production systems: micro-level evidence from Kenya. Agriculture & Food Security, 7, 80 doi:10.1186/s40066-018-0230-0

Zhang Y, McCarl B and Jones J 2017 An Overview of Mitigation and Adaptation Needs and Strategies for the Livestock Sector. Climate, 5, 95. doi:10.3390/cli5040


Received 25 February 2020; Accepted 9 March 2020; Published 1 April 2020

Go to top